Project Summary/Abstract Racial differences in the utilization of therapeutic procedures in stroke patients are pervasive and contribute to worse outcomes in minorities compared to whites after stroke. Swallowing dysfunction immediately after stroke is common, but most patients recover adequate swallowing function within a few days or weeks. Percutaneous gastrostomy (PEG) tubes for enteral feeding are commonly placed for patients who are perceived unlikely to recover adequate swallowing function in a timely manner, but ideally are avoided as PEG placement is invasive, carries the risk of procedure-related complications, and has been associated with poor long-term outcomes. Clinical decision-making surrounding PEG placement lacks objective and standardized criteria, but is largely dependent on clinical swallowing evaluations by Speech-Language Pathology (SLP) providers, and use of ancillary radiographic swallow studies such as a videofluoroscopic swallow study (VFSS). The PI?s preliminary data show that PEG tubes are overutilized in minorities compared to whites despite similar stroke severity, medical comorbidities, and incidence of swallowing dysfunction. The research proposed in the current application will investigate the potential mechanisms underlying racial disparities in PEG tube placement after stroke by thoroughly addressing the various care processes involved in evaluation of swallowing recovery and decision-making regarding PEG placement. The specific aims are (1) to investigate whether access to SLP services including frequency of bedside swallow evaluations and VFSS testing differs among black and white stroke patients, (2) to determine the presence of an unconscious (implicit) racial bias and its effect on PEG decision-making among SLP providers, and (3) to assess whether patient-SLP provider communication behavior during bedside swallow evaluations differs between race-concordant (white patient, white SLP provider), and race-discordant (black patient, white SLP provider) patient-SLP provider pairs. This project is proposed as part of a career development plan for the PI, a cerebrovascular neurologist, to obtain training in methodology related to the study of implicit bias, patient-provider communication, and advanced statistical methods. The training program and the mentoring team the PI has assembled will help to elucidate important mechanisms underlying racial disparities in PEG placement after stroke, position him to apply the newly acquired skill set to investigate implicit bias and patient-provider communication behavior as the underlying mechanisms of other disparities in stroke care, and allow him to develop strategies aimed at mitigating disparities in stroke care via future interventional studies as he transitions to become an independent clinical investigator.